IJSPT

ORIGINAL RESEARCH

A DELPHI STUDY OF RISK FACTORS FOR ACHILLES TENDINOPATHY- OPINIONS OF WORLD TENDON EXPERTS Seth O’Neill, MSc, BSc, PGCE HE, MSCP, MACP1 Paul J. Watson, PhD, PGCE HE, FCSP2 Simon Barry, PhD, PGCE HE, MCSP1

ABSTRACT Background and Purpose: Achilles tendinopathy can be a debilitating chronic condition for both active and inactive individuals. The identification of risk facors is important both in preventing but also treating tendinopathy, many factors have been proposed but there is a lack of primary epidemiological data. The purpose of this study was to develop a statement of expert consensus on risk factors for Achilles tendinopathy in active and sedentary patient populations to inform a primary epidemiological study. Study design: Delphi study Methods and Measures: An online Delphi study was completed inviting participation from world tendon experts. The consensus was developed using three rounds of the Delphi technique. The first round developed a complete list of potential risk factors, the second round refined this list but also separated the factors into two population groups – active/athletic and inactive/sedentary. The third round ranked this list in order of perceived importance. Results: Forty-four experts were invited to participate, 16 participated in the first round (response rate 40%) and two dropped out in the second round (resulting in a response rate of 35%). A total of 27 intrinsic and eight extrinsic risk factors were identified during round one. During round two only 12 intrinsic and five extrinsic risk factors were identified as important in active/athletic tendinopathy while 14 intrinsic and three extrinsic factors were identified as important for inactive/sedentary tendinopathy. Conclusions: Risk factors for Achilles tendinopathy were identified based on expert consensus, and these factors provide a basis for primary epidemiological studies. Plantarflexor strength was identified as the primary modifiable factor in the active/athletic group while systemic factors were identified as important in the inactive/sedentary group, many of the potential factors suggested for either group were non-modifiable. Non-modifiable factors include: previous tendinopathy, previous injury, advancing age, sex, steroid exposure, and antibiotic treatment. Level of evidence: Level V Key words: Achilles tendinopathy, Delphi study, risk factors

1 2

University of Leicester, Leicester, UK Coventry University, Coventry, UK

CORRESPONDING AUTHOR Seth O’Neill, MSc, BSc, PGCE HE, MSCP, MACP Physiotherapy Lecturer Department of Medical and Social Care Education Maurice Shock Medical Sciences Building G77, University of Leicester PO Box 138, LE1 9HN Tel 0116 252 3305 E-mail [email protected]

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 684

INTRODUCTION Achilles tendinopathy (AT) is a chronic potentially debilitating condition that can persist for years, preventing people from participating in physical activities like walking and running. It affects 2% of the general population1 and may have a lifetime prevalence of 42% in more active groups2 with most authors reporting 7-12% point prevalence rates in active/athletic individuals.3,4 While the current data supports a higher prevalence in active groups there is a suggestion that only 35% of AT cases presenting to General Practitioner’s (GP’s) are linked to sports participation.1 This apparent discrepancy may be related to sporting individuals bypassing their GP’s and seeking care outside of the normal health care system or may be the result of GP’s failing to identify or seek appropriate sporting activities and hobbies during a subjective examination. An important aspect of risk factor analysis is the understanding etiology of tendinopathy. The most accepted theoretical patho-etiological model is the “continuum model”,5 although others exist.6,7 The common feature of these models is excessive loading causing a loss of tissue homeostasis8 although the iceberg model6 and Fu et al7 suggest inflammatory involvement and failed healing while the continuum model fails to accept inflammatory mediated reactions. Instead the continuum model proposes a cell mediated reaction devoid of inflammatory components.5,9 This effectively leads to an active process of degenerative change involving pathways associated with inflammation.10-12 This “cell mediated reaction” is a biochemical cascade leading to alterations in structure and function of the affected tendon.5,9,13 Essentially, tendinopathy seems to link to the rate of wear being greater than the rate of repair.13,14 This simple theory can be used to hypothesise risk factors into two groups; those that increase wear, and those that limit the repair process. An example disorder affecting the ability of the tendon to repair is diabetes. Diabetes is thought to alter the glycation of collagen within tendons affecting their structural integrity, thereby reducing the capacity of the tendon to tolerate load,15 whereas training loads (distance, intensity, duration) influence the rate of wear on the tissue causing some transient degradation in collagen content.16,17 In vivo human research supports these models and confirms a net degradation

in collagen after loading.6-18 The current literature also suggests a split in risk factors between the two common clinical groups – inactive/sedentary versus active/athletic individuals. The active group may be more influenced by extrinsic factors such as training errors while the inactive/sedentary group is thought to be influenced by intrinsic factors that are often systemic e.g. inflammatory arthropathies or diabetes mellitus. In recent years there has been an increase in studies examining risk factors for AT, the majority of these are cross sectional, often termed association studies19-28 rather than longitudinal studies, which are often prospective in nature.29-34 The two types of studies have different uses. Association studies develop information about relationships between diseases and variables whilst prospective studies allow some measure of the cause and effect relationship to be established. The current literature suggests that risk factors for the development of AT may include: sex,35-37 advancing age,14,35,38,39 obesity,33,38,40,41 diabetes mellitus,42 genetics,43 high cholesterol,38,44 hypertension,42,45 tendon structure,31 hormone replacement therapy,46,47 the use of female contraceptive pills,48 early menopause,47 steroid and antibiotic exposure,38,49 recent injuries or previous tendinopathy,50 foot pronation,51 dorsiflexion range of motion,34,52,53 altered gait kinematics and kinetics,54 rheumatological disease, 55 previous sciatica, “training errors” and alterations to activity levels, both higher and lower,50 footwear,20 environmental factors like temperature,56 training surfaces,57,58 and muscle strength.29,34 Unfortunately many of these factors are not based on primary epidemiological data and often represent author opinion. Two systematic reviews exist in the area with one addressing biomechanical alterations54 and the other focusing solely on runners.59 Neither of these reviews identified many strong risk factors, as the evidence is weak for most variables. Only prospective studies can be used to make conclusions about causality, unfortunately there are few prospective studies and of those that exist seem to have used a “fishing” approach to the variables examined. This has led to the identified variables lacking plausible biological explanations. Many of the reported variables relate only to the individual researcher’s beliefs and assumptions about the

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 685

METHOD

Participants Purposive sampling was used to identify world’s leading tendon experts who had published at least two research papers on AT in the prior 10 years. Potential participants were from a clinical background and included Sports medicine physicians, surgeons or physical therapists. The potential participants were identified from published articles by using a search involving Achilles tend* as this identifies all versions of tendon, (e.g. tendinitis, tendinosis, tendonosis, tendinopathy etc) on electronic databases including CINAHL, Medline, Sports Discuss, and Scopus. In total 44 world tendon experts were identified as suitable participants. The majority of these experts were highly influential academics with long career histories. Contact information was extracted from publications and initial emails sent to determine interest. Informed consent was obtained and the rights of individuals were preserved. In order to maintain anonymity for the experts in this study they have not been named, in an attempt to show their academic “quality” we have provided information on their H index based on the 10-year period prior to the study. The H index is a measure of impact of individual academics. The numerical ranking expresses how many articles they have produced that have been cited that number of times. This means that an H index of 10 equates to ten publications with ten citations each. The Achilles tendon experts included in this study had H index ranging from 3-66 when assessed using the Scopus database. Isolating the H index of the participants to only their tendon papers reduces the H number, with the highest-ranking author having an index of 19 and the lowest ranking participant an index of 2. Ten of the participants ranked within the top 60 Achilles tendon experts identified by www.expertscape.com, a website which ranks clinical and academic medical professionals based on their publications. It is important to note that many of the academics on the Expertscape website did not meet the participant inclusion criteria as they were basic scientists and not clinicians. Using the data from the subjects H index and Expertscape ranking it is clear that the study successfully recruited world leading Achilles tendon experts.

Study design This study used a Delphi design of world tendon experts. Ethical approval was given through the University of Leicester ethics review panel.

Procedure An initial literature review was completed to identify potential risk factors; these factors were formulated

etiology of AT, although occasionally the studies are informed by previous research findings. Although a systematic review is an appropriate method to identify the current state of evidence it has a limited value in examining current theories and informing future directions as it can only report the current literature. The lack of identification of new risk factors for AT has caused some stagnation in the list of risk factors being investigated, in particular with regard to preventative studies.60 Because of the inconclusive findings and lack of full and total consideration of all possible risk factors further research is needed to address this gap. In areas such as AT where there is limited literature/evidence substantiating risk factors it is appropriate to develop an expert consensus. The development of expert consensus is best done using a Delphi technique as all members of the panel can offer their opinion without hierarchical issues affecting decisions.61 A Delphi technique involves the collection of expert opinion using structured or semi-structured rounds of questioning leading to a consensus of opinion, particularly useful for informing clinical decision-making.62-64 The views of world experts were deemed more appropriate than those of informed individuals as the experts are those publishing the studies whose findings are being disseminated to the “informed individuals”. These same experts are influencing others research agendas, either through reviewing grant applications, peer review of publications, conference presentations, or simply through their own research teams. Effectively these experts set the “trend” for further research and as such they provide an important group whose opinions are worthy of examination. The purpose of this study was to develop a statement of expert consensus on risk factors for AT in active and sedentary patient populations to inform a primary epidemiological study. This was to include previously identified factors and also new possibilities. The identified factors could then be used to inform further epidemiological research.

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 686

Risk Factors for Achilles Tendinopathy- Opinions of World Tendon Experts

First Round

Parcipants (N=44) invited by email.

Parcipants (N=16) assessed the suggested risk factors and added any other factors

Due to the ambiguity of the term “Training errors” experts who idenfied this risk factor were asked to specify components they considered e.g. training distance, alteraons in training etc.

Second Round

Parcipants were emailed the results of the first round

Parcipants (N=14) reviewed the results and idenfied which factors were relevant for two disnct paent populaons - Acve/Athlec individuals and Inacve/Sedentary

During the second round respondents idenfied muscle power as one of the few modifiable factors and as many of the intrinsic and extrinsic factors may be confounded by strength or muscle endurance it was deemed appropriate to ask if respondents considered strengthening exercises of the Triceps Surae to be preventave of Achilles tendinopathy

Risk factors that were idenfied by less than 40% of respondents were removed from the next round

Third Round

Parcipants were emailed the results of the second round

Parcipants ranked risk factors in priority order for each of the disnct paent populaons Acve/Athlec individuals and Inacve/Sedentary

Results of the third round were reported back to respondents

Figure 1.

into the first Delphi round. A flow chart of the procedure can be seen in Figure 1. This review identified factors from previous association and prospective cohort studies. In total, 17 intrinsic and eight extrinsic risk factors were identified in the literature. This

list was distributed to all 44 identified experts via an online survey tool (Qualtrics). Four emails were returned “unable to deliver” and these four experts were removed from the panel list. The survey ran from October 2012 to January 2013.

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 687

The first round asked respondents to identify relevant factors from the disseminated list that they felt were of importance, and they were asked to add any other factors they felt were important but not already listed. These additional factors were then added to the existing list and used for round two. The ability of experts to identify any other factors omitted from the first round is highly recommended for Delphi studies and helps ensure a valid complete set of factors.65 A commonly identified extrinsic risk factor in previous studies was “training errors”, however there is no clear operational definition of this term. In an attempt to develop knowledge around factors considered as “training errors” those participants who identified “training errors” as a factor were directed to a further question determining which specific components they considered important, this included: weekly distance, years in sport, recent alterations in training, change in type of activity, alterations in amount of activity, or accumulation of time active/inactive. Round two asked participants to identify which factors from the full list they felt were applicable. In addition to this information the study sought to distinguish risk factors for two distinct groups- an Active/ Athletic group and an Inactive/Sedentary group. For the purposes of this study we defined an: Active/ athletic individual as someone who participates in vigorous physical activity a minimum of three times per week. Conversely, inactive/sedentary individual were defined as those individuals who do not participate in physical activity or hobbies. Factors not identified by 40% or more of the panel as applicable were removed, in keeping with the Delphi technique.65,66 During the second round respondents identified muscle power as one of the few modifiable factors and as many of the intrinsic and extrinsic factors may be confounded by strength or muscle endurance it was deemed appropriate to ask whether respondents considered strengthening exercises of the Triceps Surae to be preventative of AT. This question was asked for both the Active/Athletic and Inactive/Sedentary groups. Respondents were asked to justify their answer to this question for both populations during the second round. The third round focussed on ranking in priority order the factors experts considered most important.

This round involved intrinsic and extrinsic factors and was ranked separately for the two populations active/athletic and inactive/sedentary. Data Analysis and feedback In between each round respondents were provided with the results of the previous round. This is recommended in Delphi technique as it helps participants consider their response in light of their peers. 65-68 However the anonymous nature of this feedback helps to reduce bias often inadvertently caused by perceived hierarchical roles.66 The data was analysed using simple descriptive methods. Email reminders were sent to participants in an attempt to increase participation levels. A limit was set at two email requests so that the respondents did not feel inappropriately pressured,69,70 as required by the ethics approval and is recommended as a valid way of increasing response rates. Each round ran for a total of two weeks. RESULTS Participants Round 1 was comprised of 16 participants from the initial 40 who were contacted, a response rate of 40%. All 16 respondents were active clinical researchers eight physical therapists, five consultant surgeons, two Sports Medicine Physicians, and one Exercise Physiologist. The second and third rounds had 14 participants, a response rate of 35%. It is unclear why two individuals were lost to follow up between rounds 1 and 2 as no response was forthcoming to the reminder emails. The two drop outs were one physical therapist and one consultant surgeon. Round 1 Round 1 started with 17 Intrinsic and eight extrinsic factors, and a further 10 intrinsic factors were identified by respondents. If respondents ticked “training errors” they were asked another question seeking to identify what elements they considered within this category, this also allowed further choices to be added. The “training errors” had six components initially proposed with a further three added by the experts. Many of these factors could be considered overlapping in nature. These factors were not included in the later rounds as they were under the umbrella term of “training errors”.

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 688

Round 2 Round 2 started with 27 intrinsic factors and eight extrinsic factors, which were then split into the active and inactive groups. All factors were included for both AT populations but experts considered the factors independently for each group. The results are presented separately for both the populations considered. Round 2 - Active/Athletic In the active/athletic group only 12 intrinsic and five extrinsic factors were identified by more than 40% of panel and therefore included in round 3, the other factors were removed from the study. Round 3 ranked the factors in order using the 12 Intrinsic factors and five extrinsic factors, and within the extrinsic factors “training errors” were further split. This was an attempt to develop further understanding about what is considered a training error. Round 2- Inactive/Sedentary In round 2 the inactive/sedentary group had 14 of the 27 intrinsic factors identified by 40% or more of panel, these factors then formed the basis for round 3. Of the seven extrinsic factors used in round 2 only

three were identified by 40% or more and therefore used in round 3. In this group instead of “training errors” “activity levels” were identified as an issue, due to the lack of definition further clarification was sought with sub categorization into seven factors. Round 3 Ranked results are shown for the active/athletic groups split into intrinsic (Table 1) and extrinsic (Table 3) risk factors, while the inactive/sedentary groups intrinsic and extrinsic factors are ranked in Tables 2 and 4 respectively. The extrinsic factor of “training error” for active/athletic tendinopathy and “activity levels” for the inactive/sedentary group were further categorized and ranked, and are displayed in Table 5 for the active/athletic group and Table 6 for the inactive/sedentary group. DISCUSSION This Delphi study explored risk factors for AT. The experts who participated in this study suggested that muscle strength/weakness is the primary modifiable risk factor for tendinopathy in an active/ athletic population. They also suggested muscle strengthening may be an important preventative

Table 1. Ranked Intrinsic risk factors for the development of Achilles tendinopathy in active/athletic individuals

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 689

Table 2. Ranked Intrinsic risk factors for the development of Achilles tendinopathy in inactive/sedentary individuals

Table 3. Ranked Extrinsic risk factors for the development of Achilles tendinopathy in Active/Athletic individuals.

Table 4. Ranked extrinsic risk factors for the development of Achilles tendinopathy in inactive/sedentary individuals.

measure. This clearly links to current therapeutic management and highlights an area where focussed research is needed. The experts suggested that many in the Inactive/sedentary group have factors that are related to systemic changes (age, sex, adiposity, and diabetes). These findings are discussed below. Intrinsic Risk Factors for Active/Athletic Tendinopathy The intrinsic risk factors for the athletic group can be seen in Table 1, the first two relate to previous

injuries and are not directly modifiable, however a recent meta-analysis showed that strength training could reduce overuse injuries in the lower limb by 50% and acute injuries by 33%.71 Therefore the integration of strength training in athletes may well assist modifying these two factors. Strength training was the highest-ranking modifiable factor identified by the experts in this panel. Currently strength training appears to be the mainstay of therapeutic exercise for tendinopathy and the authors of a recent systematic review suggested that neuromuscular adapta-

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 690

Table 5. A ranked breakdown of what the panel considered as a “training error” for the Active/Athletic individuals.

Table 6. A ranked breakdown of what the panel considered within the “activity level” risk factor for the inactive/sedentary individuals.

tions are the only consistent potential mechanism of effect.72 Current data also shows that if rehabilitation from previous tendinopathy is incomplete leaving neuromuscular deficits (functional performance, strength and endurance)73-75 these residual strength deficits may explain why previous injury is considered such a risk factor for recurrent tendinopathy. 76 Alternatively the incomplete resolution of structural changes may explain the risk.77 The third and fourth ranked intrinsic risk factors, advancing age and sex, are not amenable to change, however the link between advancing age and sex and the development of AT may simply be a confounder, with the real issue being age related reductions in strength, rather than age specifically.78 Age and sex are often mentioned as risk factors for tendinopathy but this has only been studied once,35 and no link was found between age or sex and tendinopathy.35 There is also a suggestion that any between sex differences may simply be the result of differences in reporting of injuries or seeking care.79 This discrepancy between the expert consensus and some of the literature highlights the complex nature of risk factors and the difficulty analyzing the literature.

Intrinsic Risk Factors for Inactive/Sedentary Tendinopathy The intrinsic risk factors identified for the inactive/ sedentary group (table 2) differ from the active/ athletic group (Table 1). Many of the high-ranking factors in the inactive/sedentary group represent systemic elements. These are all theorized to alter the tissue homeostatic mechanisms leading to reduced capacity to tolerate load.80 The experts in the study identified weight and obesity as separate factors. The rationale for this is unclear because further information was not sought. However the current literature suggests obesity is linked to an increase in pro-inflammatory cytokines and this is hypothesized to influence tissue homeostasis by influencing repair rates, which seems to be particularly relevant to abdominal adipose tissue. Weight per se is thought to influence the mechanical load on the tissue and therefore is linked to the rate of wear, however, an individual can have a high weight but be non-obese, as is seen in muscular athletic individuals. Some individuals may be the reverse of this - normal weight but large amounts of adipose tissue, specifically around the waist. The reason obe-

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 691

sity and weight did not feature so highly in the athletic group is that the panel likely perceived most active/athletic individuals to be relatively slim. Many of the factors identified in the sedentary group could be perceived as metabolic syndrome and possibly grouped together. Metabolic syndrome is the combination of abdominal obesity, high cholesterol, diabetes mellitus or insulin resistance and hypertension,81 it is common that individuals have all of these factors concurrently and each factor may not be independent of the other and therefore best considered as one. Further work is needed to consider this supposition. Extrinsic Factors The panel identified similar extrinsic risk factors for both the active/athletic group and the inactive/sedentary group. The majority of factors were linked to tendon loading and were considered under several guises, with changes in loading being consistently ranked first, while more general activity levels came second. Many of the factors identified by the expert panel could be interchangeable between the two groups or indeed amongst themselves, for example, changes in loading could be considered a training error. However these factors were identified by the panel and as such were maintained as separate entities. The “training errors” and “activity level” factors were further separated and the expert panel identified sub-components as the primary factors within these, as displayed in Tables 5 and 6.The relationship between tendinopathy and load has been well documented in the literature and seems to be most commonly associated with active/athletic tendinopathy where the load is quantifiable13,79,80 and alterations to reduce load appear to reduce tendinopathy.17 Unfortunately, there is very little literature on loading in AT in inactive/sedentary populations. Does Strength Training Protect Against Achilles Tendinopathy? Physical activity was identified as a key risk factor during the Delphi study and many of the identified elements may link with muscle strength and endurance capacity being pivotal in AT development. Due to this suggested role of muscle function (strength/endurance), it was deemed appropriate to seek information from the experts about whether

Table 7. Experts responses to whether they considered strength training of the Triceps Surae group as preventative for the development of Achilles tendinopathy in Active/ Athletic patients.

Table 8. Experts responses to whether they considered strength training of the Triceps Surae group as preventative for the development of Achilles tendinopathy in Inactive/ Sedentary patients?

they believed strengthening of the plantarflexors was considered protective against AT (Tables 7 and 8). Seventy-nine percent of the panel agreed that strengthening the plantarflexors in the inactive/ sedentary group (Table 8) would prevent tendinopathy, despite ranking strength/power as the 11th risk factor. In contrast strength/power was ranked 5th in the active/athletic group but only 57% of the panel thought strength training would be preventative (Table 7). Using the panel’s stated rationale for their decisions it appears they perceived athletic individuals to already have good strength while the inactive/sedentary group were perceived as weaker. This perception of the experts seems to assume that because individuals are active they are inherently strong enough for their chosen activity. Currently several authors have concluded that this is not the

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 692

case and clinicians should consider strength in relation to body weight and functional requirements rather than just gross strength.75,82,83 Many of the identified intrinsic and extrinsic factors seem capable of inducing muscle weakness through either over training – “training errors”/”activity modification” or under training/inactivity. Over training could induce muscle weakness by allowing insufficient recovery periods between bouts of activity, studies have shown muscle weakness/fatigue may last for up to 47 days after a single session of exercise. 84 Many of the individuals in this study reported feeling “recovered” despite ongoing neuromuscular (strength, endurance and functional) deficits. It would appear likely than many individuals experience the prolonged neuromuscular deficits with their normal training regimes but fail to acknowledge this residual weakness/fatigue84,85 and inadvertently continue training, thereby increasing the risk of tendinopathy. Undertraining/inactivity would have a similar effect on strength although through other various neuromuscular processes including atrophy. The current literature on adiposity/obesity has focused on systemic changes linked to cytokine levels. It is possible that obesity, particularly abdominal adiposity, increases the stress and strain on the Achilles tendon. This would occur as a consequence of abdominal adiposity moving the centre of mass forwards thereby increasing the load on the Achilles tendon. This alteration in Achilles tendon stress and strain may lead to greater negative tendon adaptation and consequently to tendinopathy. The role muscle strength plays in AT is poorly researched, however, data on army recruits shows that plantarflexor strength is predictive of AT development,34 and plantarflexor weakness has been repeatedly associated with AT.19,23 However it is unknown whether modifying strength alters the risk for tendinopathy, or how it may link to clinical outcome.19,34 Further studies are needed to examine whether strength training is preventative of tendinopathy, as well as to investigate the underlying mechanism behind clinical changes that occur with strength training (loading). The two main clinically feasible explanations for recovery are tendon adaptation,86-89 however this is not supported by current evidence in a recent systematic review ,11 or neuromuscular adaptations that help protect the tendon

from load.34,72,76,90,91 Unfortunately there is limited evidence assessing alterations in muscle function after therapeutic intervention. Limitations of findings In interpreting the findings of this study it is important to consider methodological limitations. While the sample size of 14 appears relatively small it must be considered in relation to the number of leading authorities on AT and the number of respondents in other published Delphi studies, with 17 respondents being the average.65 Since the number of world Achilles tendon experts is relatively small a sample size of 14 would have been a reasonable proportion of the overall population size. A response rate of 40% for an online survey is acceptable and of sufficient size to produce meaningful data.65 The expert panel that participated in this study may produce a different list of factors if the methodology were altered to include clinicians rather than published researchers. However this study sought expert opinion due to the lack of empirical evidence. The impact of nonresponse to the survey is difficult to determine, as no information is available for the rationale for nonresponse to the initial invitation. In order to improve participation and reduce the burden the study used relatively short rounds. The dropout rate from the start of round one to completion of the study was only two individuals, or 12% drop out rate. The reasons for dropping out are unknown. It is possible that risk factors were missing from the initial list but an exhaustive review was undertaken and the experts involved were allowed to offer alternative variables, which were then included in the second round of the Delphi. A potential problem with using ranking is many of these factors may be considered in combination with one another and not relative isolation,92 this may make clusters of factors more important in some individuals. A key limitation to the findings is that they represent the opinions of leading experts in the area, and although such findings are useful it is no substitute for carefully conducted prospective epidemiological studies. This Delphi study did not ask experts to differentiate between insertional and mid-portion tendinopathy. It may be that different factors play different roles for each of these disorders as the disease etiologies are

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 693

felt to differ e.g. insertional tendinopathy is felt to be influenced by compression93 of the Achilles tendon against the calcaneus94,95 and therefore increased dorsiflexion may play a more pivotal role. However this was not investigated as the current body of risk factor evidence has failed to distinguish between the two zones or types of tendinopathy. By developing a complete list of factors initially it is hoped that further epidemiological studies can distinguish between these two groups more specifically. Further studies The ranking of the identified factors helps develop understanding about which factors may be most important in the etiology of AT. This data will help to focus further prospective longitudinal research and also guide clinicians with factors to address during their clinical interventions. While this study has developed new knowledge around risk factors, no cause and effect can be inferred, and it is still unknown why particular factors may increase risk of tendinopathy. Further studies are needed to confirm the suggestions from world experts and then to develop further understanding regarding the underlying pathological processes. The opinions of international experts are likely to influence the way in which tendon research is conducted through the peer review process for both publication and research grant awards. Careful epidemiological research is required to substantiate these opinions to facilitate objective research. It is through this process that preventative regimes may be developed. CONCLUSION This study produces new data regarding potential risk factors for AT. World tendon experts identified the main modifiable risk factor in active/athletic group to be muscle strength, which is supported by some prospective data.34 A focus on measuring muscle strength/power/endurance in prospective cohorts to determine normal values and what level constitutes a risk appears to have merit. However in inactive/sedentary groups the focus may need to be on obesity, weight, diabetes and cholesterol levels, all of which are systemic risk factors for tendinopathy. While these factors are modifiable in nature, the ability to change these during a normal therapeutic intervention period may be challenging.

REFERENCES 1. de Jonge S, van den Berg C, de Vos R, van der Heide H, et al. Incidence of midportion Achilles tendinopathy in the general population. Br J Sports Med. 2011;45:1026-1028. 2. Kujala UM, Sarna S, Kaprio J. Cumulative incidence of Achilles tendon rupture and tendinopathy in former male elite athletes. Can J Sports Med. 2005(15):133-135. 3 . Marti B, Varder J, Minder C, Avelin T. On the epidemiology of running injuries- the 1984 Bern grand prix study. Am J Sports Med. 1985;16(3):285-294. 4. Johannessen C. Injuries in elite orienteers. Am J Sports Med. 1986;14(5):410-415. 5. Cook J, Rio E, Purdam C, Docking S. Revisiting the continuum model of tendon pathology:what is its merit in clinical practice and research? Br J Sports Med. 2016;(Epub ahead of print). 6. Abate M, Silbernagel K, Siljeholm C, Lorio A, De Amicus D, Werner S, et al. Pathogensisis of tendinopathies: inflammation or degeneration? Arth Res Ther. 2009;11(3). 7. Fu S, Cheuk Y, Lui P, Chan K. Deciphering the pathogenesis of tendinopathy: a three-stages process. Sp Med Arth Rehab Thera Tech. 2010;2(30). 8. Arnoczky S, Lavagnino M, Egerbacher M. The mechanobioligical aetiopathogensis of tendinopathy: is it the over-stimulation or the under-stimulation of tendon cells? Int J Exp Path. 2007;88:217-226. 9. Cook JL, Purdam CR. Is tendon pathology a continuum? A pathology model to explain the clinical presentation of load-induced tendinopathy. Br JSports Med. 2009;43(6):409-416. 10. Rees J, Stride M, Scott A. Time to revisit inflammation. Br J Sports Med. 2013;48(21):1553-1557. 11. Drew B, Smith T, Littlewood C, Sturrock B. Do structural changes (eg, collagen/matrix) explain the response to therapeutic exercises in tendinopathy: a systematic review. Br J Sports Med. 2014;48(12):966972. 12. Dean B, Gettings P, Dakin S, Carr A. Are inflammatory cells increased in painful human tendinopathy? A systematic review. Br J Sports Med. 2015;50(4):216-220. 13. Scot A, Docking S, Vincenzino B, Alfredson H, Zwerver J, Lundgreen K, et al. Sports and exerciserelated tendinopathies: a review of selected topical issues by participants of the second International Scientific Tendinopathy Symposium (ISTS) Vancouver 2012. Br J Sports Med 2013;47(9):536-544. 14. Maffulli N, Kenward M, Testa V, Capasso G, Regine R, King J. Clinical diagnosis of Achilles tendinopathy with Tendinosis. Can J Sports Med 2003;13(1):11-15.

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 694

15. Abate M, Schiavone C, Salini S. Neoangiogenesis is reduced in chronic tendinopathies of type 2 diabetic patients. Int J Immun and Pharm. 2012;5(3):757-761. 16. Langberg H, Skovgaard D, Petersen LJ, Bulow J, Kjaer M. Type I collagen synthesis and degradation in peritendinous tissue after exercise determined by microdialysis in humans. J Physiol (Lond.) 1999;521(Pt 1):299-306. 17. Rosengarten S, Cook J, Bryant A, Cordy J, Daffy J, Docking S. Australian football players’ Achilles tendons respond to game loads within 2 days: an ultrasound tissue characterisation(UTC) study. Br J Sports Med 2014;49(3):183-187. 18. Magnusson S, Langberg H, Kjaer M. The pathogenesis of tendinopathy: balancing the response to loading. Nat Rev Rheum 2010;6(5):262268. 19. Alfredson H, Pietila T, Josson P, Lorentzon R. Heavy load eccentric calf muscle training for the treatment of chronic Achilles tendinosis. Am J Sports Med. 1998;26(3):360-366. 20. Azevedo L, Lambert M, Vaughan C, O’Connor C, Schwellnus M. Biomechanical variables associated with Achilles tendinopathy in runners. Br J Sports Med. 2009;43(4):288-292. 21. Baur H, Divert C, Hirschmuller A, Muller S, Belli A, Mayer F. Analysis of gait differences in healthy runners and runners with chronic Achilles tendon complaints. Isok Ex Sci 2004;12(2):111-116. 22. Donoghue O, Harrison A, Laxton P, Jones R. Lower limb kinematics of subjects with chronic Achilles tendon injury during running. Res Sports Med. 2008;16(1):23-38. 23. McCrory J, Martin D, Lowery R, Cannon D, Curl W, Read H, et al. Etiologic factors associated with Achilles tendinitis in runners, Med Sci Sports Exerc. 1999;31:1374-1381. 24. Ryan M, Grau S, Krauss I, Maiwald C, Taunton J, Horstmann T. Kinematic analysis of runners with Achilles mid-portion tendinopathy. Foot & Ankle Int. 2009;12:1190-1195. 25. Taunton J, Ryan M, Clement D, McKenzie D, LloydSmith D, Zumbo B. A retrospective case-control analysis of 2002 running injuries. Br J Sports Med. 2002;36(2):95-101. 26. van der Linden P, van de Lei J, Nab H, Knol A, Stricker B.  Achilles tendinitis associated with fluoroquinolones. Br J Clin Pharm. 1999;48(3):433-437. 27. van Gent R, Siem D, van Middelkoop M, van Os A, Bierma-Zeinstra SM, Koes BW. Incidence and determinants of lower extremity running injuries in long distance runners :a Systematic Review. Br J Sports Med. 2007;41(8):469-480.

28. Williams D, Zambardino J, Banning V. Transverseplane mechanics at the knee and tibia in runners with and without a history of Achilles tendonopathy. J Orthop Sports Phys Ther.2008;38(12):761-767. 29. Hein T, Janssen P, Wagner-Fritz U, Haupt G, Grau S. Prospective analysis of intrinsic and extrinsic risk factors on the development of Achilles tendon pain inrunners. Scand J Med Sci Sports. 2013;24(3):e201-e212. 30. Kaufman K, Brodine S, Shaffer R, Johnson C, Cullison T.  The effect of foot structure and range of motion on musculoskeletal overuse injuries. Am J Sports Med. 1999;27(5):585-593. 31. Fredberg U, Bolvig L. Significance of ultrasonographically detected asymptomatic tendinosis in the patellar and achilles tendons of elite soccer players: a longitudinal study. Am J Sports Med. 2002;30(4):488-491. 32. Van Ginckel A, Thijs Y, Hesar N, Mahieu N, De Clerq D, Roosen P, et al. Intrinsic gait-related risk factors for Achilles tendinopathy in novice runners: a prospective study. Gait & Post. 2008;29(3):387-391. 33. Gaida J, Ashe M, Bass S, Cook J. Is adiposity an under-recognized risk factor for tendinopathy? A systematic review. Arthr Care Res. 2009;61(6):840849. 34. Mahieu NN, Witvrouw E, Stevens V, Van Tiggelen D, Roget P. Intrinsic risk factors for the development of achilles tendon overuse injury: a prospective study. Am J Sports Med. 2006;34(2):226-235. 35. Longo UG, Rittenberg J, Garau G, Radonic B, Gutwasser C, Gilliver SF, et al. No Influence of Age, Gender, Weight, Height and Impact Profile in Achilles Tendinopathy in Masters Track and Field Athletics. Am J Sports Med. 2009;37(7):1400-1405. 36. Jones BH, Bovee MW, Harris JM,3rd, Cowan DN. Intrinsic risk factors for exercise-related injuries among male and female army trainees. Am J Sports Med. 1993;21(5):705-710. 37. van Mechelen W. Running injuries. A review of the epidemiological literature. Sports Med. 1992;14(5):320-335. 38. Wise BL, Peloquin C, Choi H, Lane NE, Zhang Y. Impact of age, sex, obesity, and steroid use on quinolone-associated tendon disorders. Am J Sports Med. 2012;125(12):1228.e23-1228.e28. 39. Maffulli N, Wong J, Almekinders LC. Types and epidemiology of tendinopathy. Can J Sports Med. 2003;22(4):675-692. 40. Gaida JE, Alfredson L, Kiss ZS, Wilson AM, Alfredson H, Cook JL. Dyslipidemia in achilles tendinopathy is characteristic of insulin resistance. Med Sci Sports Exerc. 2009;41(6):1194-1197.

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 695

41. Addison O, LaStayo PC, Dibble LE, Marcus RL. Inflammation, aging, and adiposity: implications for physical therapists. J Geri Phys Ther. 2012;35(2):86-94. 42. Batista F, Nery C, Pinzur M, Monteiro AC, de Souza E, Felippe F, et al. Achilles tendinopathy in diabetes mellitus. Foot Ankle Int. 2008;29(5):498-501. 43. Raleigh SM, van der Merwe L, Ribbans WJ, Smith RK, Schwellnus MP, Collins M. Variants within the MMP3 gene are associated with Achilles tendinopathy: possible interaction with the COL5A1 gene. Br J Sports Med. 2009;43(7):514-520. 44. Windler E, Beil F-, Altenburg C, Rinninger F. Familial hypercholesterolemia - A common metabolic disease rarely diagnosed. Dtsch Med Wochenschr. 2012;137(46):2375-2378. 45. Holmes GB, Lin J. Etiologic factors associated with symptomatic Achilles tendinopathy. Foot Ankle Int. 2006;27(11):952-959. 46. Knobloch K, Schreibmueller L, Kraemer R, Jagodzinski M, Vogt PM, Redeker J. Gender and eccentric training in Achilles mid-portion tendinopathy. Knee Surg Sports Traumatol Arthrosc. 2010;18(5):648-655. 47. Cook J, Bass S, Black J. Hormone therapy is associated with smaller Achilles tendon diameter in active post-menopausal women. Scand J Med Sci Sports. 2007;17(2):128-132. 48. Hansen M, Couppe C, Hansen C, Skovgaard D, Kovanen V, Larsen J, et al. Impact of oral contraceptive use and menstrual phase on patellar tendon morphology, biochemical composition and biomechanical properties in female athletes. J App Phys. 2013;114(8):998-1008. 49. Sode J, Obel N, Hallas J, Lassen A. Use of fluroquinolone and risk of Achilles tendon rupture: a population-based cohort study. Eur J Clin Pharmacol. 2007;63(5):499-503. 50. Saragiotto B, Yamato T, Hespanhol L, Rainbow M, Davis I, Lopes A. What are the main risk factors for running-related injuries: A systematic review. Sports Med. 2014;44:1153-1163. 51. Van Ginckel A, Thijs Y, Hesar NG, Mahieu N, De Clercq D, Roosen P, et al. Intrinsic gait-related risk factors for Achilles tendinopathy in novice runners: a prospective study. Gait Posture. 2009;29(3):387-391. 52. Kaufman K, Brodine S, Shaffer R, Johnson C, Cullison T. The effect of foot structure and range of motion on musculoskeletal overuse injuries. Am J Sports Med. 1999;27(5):585-593. 53. Rabin A, Kozul Z, Finestone A. Limited Ankle Dorsiflexion increases the risk for mid-portion Achilles tendinopathy in infatry recruits: A prospective study. J Foot and Ankle Res. 2014;18(1):48.

54. Munteanu SE, Barton C. Lower limb biomechanics during running in individuals with Achilles tendinopathy : a systematic review. Foot and Ankle Res. 2011;4(15). 55. Jarvinen TA, Kannus P, Maffulli N, Khan KM. Achilles tendon disorders: etiology and epidemiology. Foot & Ankle Clin. 2005;10(2):255-266. 56. Milgrom C, Finestone A, Zin D, Mandel D, Novack V. Cold weather training: a risk factor for Achilles paratendinitis among recruits. Foot Ankle Int. 2003;24(5):398-401. 57. Knobloch K, Yoon U, Vogt PM. Acute and overuse injuries correlated to hours of training in master running athletes. Foot Ankle Int. 2008;29(7):671-676. 58. Di Caprio F, Buda R, Mosca M, Calabrò A, Giannini S. Foot and lower limb diseases in runners: Assessment of risk factors. J Sports Sci Med. 2010;9(4):587-596. 59. Lorimer A, Hume P. Achilles tendon injury risk factors associated with running. Sports Med. 2014;44(10):1459-1472. 60. Peters J, Zwerver J, Diercks RL, Efferink-Gemser M, Akker-Scheek I. Preventative interventions for tendinopathy: A systematic review. J Sports Sci Med. 2016;19(3):205-211. 61. Okoli C, Powlowshi S. The Delphi method as a research tool: an example, design considerations and applications. Inform Manage. 2004;42(1):15-29. 62. Powell C. The Delphi technique: myths and realities. J Adv Nursing. 2003;41:376-82. 63. Nekolaichuk C, Fainsinger R, Lawlor P. A validation study of a pain classification system for advanced cancer patients using content experts: the Edmonton classification system for cancer pain. Pall Med. 2005;19:466-476. 64. Ellis R editor. The Delphi technique: its contribution to the evaluation of professional practice. In Professional Competence and Quality Assurance in the Caring Professions. London: Chapman and Hall; 1998. 65. Boulkedid R, Abdoul H, Loustau M, Sibony O, Alberti C. Using and reporting a Delphi method for selecting healthcare quality indicators :  A systematic review. PLoS ONE [Electronic Resource] 2011;6(6):e20476. 66. Linstone H, Turoff M editors. Delphi Method: Techniques and Applications. Boston (USA): Addison-Wesley Publishing; 1975. 67. Delbecq A, Van de Ven A, Gustafson D. Group techniques for program planning: Foresman; 1975. 68. Mullen P. When is Delphi not Delphi? Discussion Paper 37 ed. UK: Health Services management centre, University of Birmingham; 2000.

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 696

69. Verhagen A, de Ve H, de Bie R, Kessels A, Boers M, Bauer L, et al. The Delphi List: A Criteria List for Quality Assessment of Randomized Clinical Trials for Conducting Systematic Reviews Developed by Delphi Consensus. J Clin Epid. 1998;51(12):1235-1241. 70. Schmidt R. Managing Delphi surveys using nonparametric statistical techniques. Decision Sciences. 1997;28(3):763-774. 71. Lauersen J, Bertelsen D, Andersen L. The effectiveness of exercise interventions to prevent sports injuries : a systematic review and metaanalysis of randomised controlled trials. Br J Sports Med. 2014;48(11):871-877. 72. Malliaras P, Barton C, Reeves N, Langberg H. Achilles and patellar tendinopathy loading programmes : a systematic review comparing clinical outcomes and identifying potential mechanisms for effectiveness. Sports Med. 2013;43(4):267-286. 73. O’Neill S, Watson P, Barry S. Eccentric exercises for Achilles tendinopathy do not fully resolve the plantarflexor muscle power deficits. Br J Sports Med. 2014;48(A49). 74. O’Neill S, Watson P, Barry S. Eccentric rehabilitation for runners with Achilles tendinopathy improves endurance capacity of the plantarflexors. Physiotherapy. 2015;101(Supplement 1):e1143-e1144. 75. Silbernagel KG, Thomee R, Eriksson BI, Karlsson J. Full symptomatic recovery does not ensure full recovery of muscle-tendon function in patients with Achilles tendinopathy. Br J Sports Med online 2007 discussion 280;41(4):276-280. 76. O’Neill S, Watson P, Barry S. Why are eccentric exercises effective for Achilles tendinopathy? Int J Sports Phys Ther. 2015;10(4):552-562. 77. Fredberg U, Bolvig L, Andersen NT. Prophylactic training in asymptomatic soccer players with ultrasonographic abnormalities in Achilles and patellar tendons: the Danish Super League Study. Am J Sports Med. 2008;36(3):451-460. 78. Opar D, Williams M, Timmins R, Hickey J, Duhiq S, Shield A. Eccentric Hamstring Strength and Hamstring Injury Risk in Australian Footballers. Med Sci Sports Exerc. 2015;47(4):857-865. 79. Almeida S, Trone D, Leone D, Shaffer P, Patheal S, Long K. Gender differences in musculoskeletal injury rates: a function of symptom reporting? Med Sci Sports Exerc. 1999;31:1807-1812. 80. Dye S. The pathophysiology of patellofemoral pain: a tissue homeostasis perspective.   Clin Orthop Rel Res. 2005;43(6):100-110. 81. Alberti K, Zimmet P, Shaw J. metabolic syndrome- a new world-wide defination. A consensus statement

82.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

from the international diabetes federation. Diabetic medicine. 2006;23(5):469-480. O’Neill S, Watson P, Barry S. Plantarflexor muscle power deficits in runners with Achilles tendinopathy. Br J Sports Med. 2014;48(A49). O’Neill S, Watson P, Barry S. Endurance deficits in patients with Achilles tendinopathy versus healthy controls. Physiotherapy. 2015;101(Supplement 1):e1142-e1143. Sayers S, Clarkson P. Force recovery after eccentric exercises in males and females. Eur J App Phys. 2001;84(1):122-126. Cleak M, Eston R. Muscle soreness, swelling, stiffness and strength loss after intense eccentric exercise. Br J Sports Med. 1992;26(4):267-272. Langberg H, Ellingsgaard H, Madsen T, Jansson J, Magnusson SP, Aagaard P, et al. Eccentric rehabilitation exercise increases peritendinous type I collagen synthesis in humans with Achilles tendinosis.  Scand J Med Sci Sports. 2007;17(1):61-66. Alfredson H, Pietila T, Jonsson P, Lorentzon R. Heavy-load eccentric calf muscle training for the treatment of chronic Achilles tendinosis. Am J Sports Med. 1998;26(3):360-366. Maffulli N, Longo UG. How do eccentric exercises work in tendinopathy? Rheumatology (Oxford) 2008;47(10):1444-1445. Stanish WD, Rubinovich RM, Curwin S. Eccentric exercise in chronic tendinitis. Clin Orthop Relat Res. 1986;(208)(208):65-68. Lindstedt S, Reich T, Keim P, LaStayo P. Do muscles function as adaptable locomotor springs? J Exp Biol. 2002;205:2211-2216. LaStayo P, Woolf J, Lewek M. Eccentric muscle contractions : their contribution to injury, prevention, rehabilitation, and sport. J Orthop Sports Phys Ther. 2003;33:557-571. Cook M, Wild C, Forman D. A systematic review and meta-analysis of the sex ratio for Barrett’s esophagus, erosive reflux disease, and nonerosive reflux disease. Am J Epid. 2005;162:1050-1061. Soslowsky L, Thomopoulos S, Esmail A, Flanagan C, Iannotti J, Williamson D, et al. Rotator cuff tendinosis in an animal model: Role of extrinsic and overuse factors. Ann Biomed Eng. 2002;30(8):1057-1063. Almekinders LC, Weinhold PS, Maffulli N. Compression etiology in tendinopathy. Clinics Sports Med 2003;22:703-710. Docking S, Samiric T, Scase E, Purdam C, Cook J. Relationship between compressive loading and ECM changes in tendons. Musc Lig Tend J. 2013;21(3):7-11.

The International Journal of Sports Physical Therapy | Volume 11, Number 5 | October 2016 | Page 697